Carbon nanotubes have emerged as promising candidates for nanoelectronic devices because of their unique electronic properties as disclosed in for example Dresselhaus, M. S., Dresselhaus, G. and Eklund, P. C. Science of Fullerenes and Carbon Nanotubes (Academic, San Diego, 1996). They exhibit either metallic or semiconducting behavior depending on the diameter and helicity of the tubes, as disclosed in Wildoer, J. W. G.; Venema, L. C.; Rinzler, A. G.; Smally, R. E.; and Dekker, C.; Nature 391, 59-61, 1998 and Odom, T. W.; Huang J.; Kim, P.; and Lieber, C. M.; Nature 391, 61-64, 1998., and can conduct current ballistically with little heat dissipation, see Frank, S.; Poncharal, P.; Wang, Z. L.; and de Heer, W. A.; Science 280, 1744-1746, 1998.
These properties have recently generated great interest in forming nanotube two-way or three-way heterojunctions between different tubes for use as building blocks in nanoelectronic devices as disclosed in Chico, L.; Crespi, V. H.; Benedict, L. X.; Louie, S. G.; and Cohen, M. L.; Phys. Rev. Lett. 76, 971-974, 1996; Mennon, M.; Srivastava, D.; Phys. Rev. Lett. 79, 4453-4456, 1997; and Mennon, M.; Srivastava, D.; J. Mater. Res., 13, 2357-2362, 1998. However, this will be difficult to achieve using conventional carbon nanotube growth methods, such as disclosed in Ebbsen, T. W., Carbon Nanotubes: Preparation and Properties (CRC Press, Boca Raton, Fla., 1997), since the straight tube structure cannot be controllably altered along its length. Ideas for post-growth modifications have been advanced in the literature, but are hard to implement in practice and prone to defects, see Collins, P. G., Bando, H. G., Zettl, A.; Nanotechnology 9, 153-157, 1998.
The ability to create devices on the scale of nanometers is a goal sought by many. One class of nanostructures which hold great promise for molecular scale devices are carbon nanotubes. In order to form electronic devices based on these structures, mainly theoretical proposals have been advanced on ways to obtain heterojunctions between different nanotubes. However, this is difficult to achieve with conventional nanotube fabrication methods, all of which produce simple straight tubes.
While there have been proposals to fabricate nanotube heterojunctions using other methods, all involve mechanically assembling or manipulating straight nanotubes after they are grown, see Collins, P. G., Bando, H. and Zettl, A. Nanotechnology 9, 153-157, 1998.
U.S. Pat. No. 5,753,088 issued to Olk is directed to a method for making carbon nanotubes. The method involves submerging carbon anode and cathode electrodes into liquid nitrogen, helium or hydrogen and passing a direct current between the electrodes thereby growing the nanotubes on the surface of the cathode.
U.S. Pat. No. 5,424,054 issued to Bethune et al. discloses a method of producing carbon fibers or tubes having a wall thickness equal to a single layer of carbon atoms. The method uses arc discharge between a carbon rod cathode and a hollowed out anode containing cobalt catalyst/carbon powder. Discharge takes place in an inert atmosphere.
U.S. Pat. No. 5,830,326 issued to Lijima teaches a method of producing carbon nanotubes using direct current discharge between carbon electrodes in a rare gas atmosphere, preferably argon.
U.S. Pat. No. 5,747,161 issued to Lijima is very similar in terms of the disclosure to Lijima ('326) discussed immediately above but the claims are directed to the product of the growth process.
U.S. Pat. No. 5,413,866 issued to Baker et al. is directed to carbon filaments produced using a thermal gas phase growth process in which a carbon containing gas is decomposed in the presence of a catalyst coated substrate. The type of metal catalyst employed has an effect on the structure of the carbon filament produced.
U.S. Pat. No. 5,457,343 issued to Ajayan et al. discloses carbon nanotubes containing foreign materials, in other words a carbon nanotube used as a storage device. The nanotubes are produced in an inert atmosphere in an electric discharge.
U.S. Pat. No. 5,489,477 issued to Ohta et al. is directed to a method of producing high-molecular weight carbon materials incorporating C.sub.60 fullerene structures.
It would be very advantageous to provide a method for producing Y-junction carbon nanotubes which can be used to construct nanoscale electronic components.